Get insights into the relation between CMV and atherosclerosis PowerPoint Presentation

Slide 1 -

Get insights into the relation between CMV and atherosclerosis Case study 1

Slide 2 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown.

Slide 3 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process.

Slide 4 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis

Slide 5 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation

Slide 6 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge

Slide 7 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!!

Slide 8 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand

Slide 9 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy

Slide 10 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes

Slide 11 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining

Slide 12 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes

Slide 13 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists

Slide 14 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions

Slide 15 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation

Slide 16 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation

Slide 17 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis

Slide 18 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs.

Slide 19 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems

Slide 20 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!!

Slide 21 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction

Slide 22 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools

Slide 23 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results

Slide 24 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model

Slide 25 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model +

Slide 26 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target

Slide 27 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array.

Slide 28 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context.

Slide 29 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - -

Slide 30 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect

Slide 31 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV.

Slide 32 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist

Slide 33 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computational modeling knowledge

Slide 34 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computational modeling knowledge Research plan Define CMV measurement method Determine plaque formation quantification method Formalize epidemiological study Perform study in population

Slide 35 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computational modeling knowledge Research plan Define CMV measurement method Determine plaque formation quantification method Formalize epidemiological study Perform study in population Research plan Local vs. global measurements CMV measurement Plaque measurement Local measurements pinpoint the exact location of plaque and CMV (= histological) Global measurements (=systemic) indicate the total amount of plaque and CMV. It is less specific than local measurements.

Slide 36 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computational modeling knowledge Research plan Define CMV measurement method Determine plaque formation quantification method Formalize epidemiological study Perform study in population Research plan Local vs. global measurements CMV measurement Plaque measurement Local measurements pinpoint the exact location of plaque and CMV (= histological) Global measurements (=systemic) indicate the total amount of plaque and CMV. It is less specific than local measurements. CMV measurement Methods used in literature CMV specific IgG antibodies Measure virus activity Global measurement Virus detection with PCR Measures the virus presence locally Do not measure the activity Done in vitro

Slide 37 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computational modeling knowledge Research plan Define CMV measurement method Determine plaque formation quantification method Formalize epidemiological study Perform study in population Research plan Local vs. global measurements CMV measurement Plaque measurement Local measurements pinpoint the exact location of plaque and CMV (= histological) Global measurements (=systemic) indicate the total amount of plaque and CMV. It is less specific than local measurements. CMV measurement Methods used in literature CMV specific IgG antibodies Measure virus activity Global measurement Virus detection with PCR Measures the virus presence locally Do not measure the activity Done in vitro Local CMV measurement No existing technique available Plan to develop measurement method: Find or develop a label that indicates CMV Attach radioactive or physical marker to the label Quantify and localise agent with nuclear (radioactive) or possibly magnetic resonance (physical) imaging technique A disadvantage of nuclear imaging is the exposure to radioactivity

Slide 38 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computational modeling knowledge Research plan Define CMV measurement method Determine plaque formation quantification method Formalize epidemiological study Perform study in population Research plan Local vs. global measurements CMV measurement Plaque measurement Local measurements pinpoint the exact location of plaque and CMV (= histological) Global measurements (=systemic) indicate the total amount of plaque and CMV. It is less specific than local measurements. CMV measurement Methods used in literature CMV specific IgG antibodies Measure virus activity Global measurement Virus detection with PCR Measures the virus presence locally Do not measure the activity Done in vitro Local CMV measurement No existing technique available Plan to develop measurement method: Find or develop a label that indicates CMV Attach radioactive or physical marker to the label Quantify and localise agent with nuclear (radioactive) or possibly magnetic resonance (physical) imaging technique A disadvantage of nuclear imaging is the exposure to radioactivity Plaque formation measurement (1) Not described or used in CMV-atherosclerosis studies It is assumed that there is no way to measure directly plaque growth Plaque growth can be determined by 2 separate plaque measurements Several possibilities are available to measure the amount of plaque locally (i.e. make an image of a subject that indicates plaque presence at each anatomical location)

Slide 39 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computational modeling knowledge Research plan Define CMV measurement method Determine plaque formation quantification method Formalize epidemiological study Perform study in population Research plan Local vs. global measurements CMV measurement Plaque measurement Local measurements pinpoint the exact location of plaque and CMV (= histological) Global measurements (=systemic) indicate the total amount of plaque and CMV. It is less specific than local measurements. CMV measurement Methods used in literature CMV specific IgG antibodies Measure virus activity Global measurement Virus detection with PCR Measures the virus presence locally Do not measure the activity Done in vitro Local CMV measurement No existing technique available Plan to develop measurement method: Find or develop a label that indicates CMV Attach radioactive or physical marker to the label Quantify and localise agent with nuclear (radioactive) or possibly magnetic resonance (physical) imaging technique A disadvantage of nuclear imaging is the exposure to radioactivity Plaque formation measurement (1) Not described or used in CMV-atherosclerosis studies It is assumed that there is no way to measure directly plaque growth Plaque growth can be determined by 2 separate plaque measurements Several possibilities are available to measure the amount of plaque locally (i.e. make an image of a subject that indicates plaque presence at each anatomical location) Plaque formation measurement (2) Plaque quantification techniques

Slide 40 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computational modeling knowledge Research plan Define CMV measurement method Determine plaque formation quantification method Formalize epidemiological study Perform study in population Research plan Local vs. global measurements CMV measurement Plaque measurement Local measurements pinpoint the exact location of plaque and CMV (= histological) Global measurements (=systemic) indicate the total amount of plaque and CMV. It is less specific than local measurements. CMV measurement Methods used in literature CMV specific IgG antibodies Measure virus activity Global measurement Virus detection with PCR Measures the virus presence locally Do not measure the activity Done in vitro Local CMV measurement No existing technique available Plan to develop measurement method: Find or develop a label that indicates CMV Attach radioactive or physical marker to the label Quantify and localise agent with nuclear (radioactive) or possibly magnetic resonance (physical) imaging technique A disadvantage of nuclear imaging is the exposure to radioactivity Plaque formation measurement (1) Not described or used in CMV-atherosclerosis studies It is assumed that there is no way to measure directly plaque growth Plaque growth can be determined by 2 separate plaque measurements Several possibilities are available to measure the amount of plaque locally (i.e. make an image of a subject that indicates plaque presence at each anatomical location) Plaque formation measurement (2) Plaque quantification techniques Epidemiological study Objective : Determine if a relation exists between CMV and plaque growth Two different possibilities Global level of CMV versus global amount of plaque growth Local level of CMV versus local amount of plaque formation CMV measurement Plaque measurement CMV level Global Local Total plaque progress Local Global

Slide 41 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computational modeling knowledge Research plan Define CMV measurement method Determine plaque formation quantification method Formalize epidemiological study Perform study in population Research plan Local vs. global measurements CMV measurement Plaque measurement Local measurements pinpoint the exact location of plaque and CMV (= histological) Global measurements (=systemic) indicate the total amount of plaque and CMV. It is less specific than local measurements. CMV measurement Methods used in literature CMV specific IgG antibodies Measure virus activity Global measurement Virus detection with PCR Measures the virus presence locally Do not measure the activity Done in vitro Local CMV measurement No existing technique available Plan to develop measurement method: Find or develop a label that indicates CMV Attach radioactive or physical marker to the label Quantify and localise agent with nuclear (radioactive) or possibly magnetic resonance (physical) imaging technique A disadvantage of nuclear imaging is the exposure to radioactivity Plaque formation measurement (1) Not described or used in CMV-atherosclerosis studies It is assumed that there is no way to measure directly plaque growth Plaque growth can be determined by 2 separate plaque measurements Several possibilities are available to measure the amount of plaque locally (i.e. make an image of a subject that indicates plaque presence at each anatomical location) Plaque formation measurement (2) Plaque quantification techniques Epidemiological study Objective : Determine if a relation exists between CMV and plaque growth Two different possibilities Global level of CMV versus global amount of plaque growth Local level of CMV versus local amount of plaque formation CMV measurement Plaque measurement CMV level Global Local Total plaque progress Local Global Epidemiological study (2) Global CMV vs. global plaque Method: Three plaque measurements with MR imaging in carotid arteries combined with automatic plaque quantification Frequent CMV activity measurements in blood samples Population: a big random selection Determine number of necessary subjects from estimation of the probability to find CMV in plaque Determine: - Difference in plaque growth between the two periods (2plaque) - Amount of times the virus was activated (NCMV) Regression : 2plaque ~NCMV AbCMV

Slide 42 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computational modeling knowledge Research plan Define CMV measurement method Determine plaque formation quantification method Formalize epidemiological study Perform study in population Research plan Local vs. global measurements CMV measurement Plaque measurement Local measurements pinpoint the exact location of plaque and CMV (= histological) Global measurements (=systemic) indicate the total amount of plaque and CMV. It is less specific than local measurements. CMV measurement Methods used in literature CMV specific IgG antibodies Measure virus activity Global measurement Virus detection with PCR Measures the virus presence locally Do not measure the activity Done in vitro Local CMV measurement No existing technique available Plan to develop measurement method: Find or develop a label that indicates CMV Attach radioactive or physical marker to the label Quantify and localise agent with nuclear (radioactive) or possibly magnetic resonance (physical) imaging technique A disadvantage of nuclear imaging is the exposure to radioactivity Plaque formation measurement (1) Not described or used in CMV-atherosclerosis studies It is assumed that there is no way to measure directly plaque growth Plaque growth can be determined by 2 separate plaque measurements Several possibilities are available to measure the amount of plaque locally (i.e. make an image of a subject that indicates plaque presence at each anatomical location) Plaque formation measurement (2) Plaque quantification techniques Epidemiological study Objective : Determine if a relation exists between CMV and plaque growth Two different possibilities Global level of CMV versus global amount of plaque growth Local level of CMV versus local amount of plaque formation CMV measurement Plaque measurement CMV level Global Local Total plaque progress Local Global Epidemiological study (2) Global CMV vs. global plaque Method: Three plaque measurements with MR imaging in carotid arteries combined with automatic plaque quantification Frequent CMV activity measurements in blood samples Population: a big random selection Determine number of necessary subjects from estimation of the probability to find CMV in plaque Determine: - Difference in plaque growth between the two periods (2plaque) - Amount of times the virus was activated (NCMV) Regression : 2plaque ~NCMV AbCMV Epidemiological study (3) Local CMV vs. local plaque Method Two plaque measurements with MR imaging in carotid arteries combined with automatic plaque quantification and localisation Image local CMV presence Computerised alignment of plaque and CMV images Population: a random selection Determine number of necessary subjects from estimation of the probability to find CMV in plaque and to find plaque Number of patients will be lower Determine: Plaque growth (plaque) at a large number of locations The presence of the virus at the same locations Regression : plaque ~ CMV CMV Local Plaque Local

Slide 43 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computational modeling knowledge Research plan Define CMV measurement method Determine plaque formation quantification method Formalize epidemiological study Perform study in population Research plan Local vs. global measurements CMV measurement Plaque measurement Local measurements pinpoint the exact location of plaque and CMV (= histological) Global measurements (=systemic) indicate the total amount of plaque and CMV. It is less specific than local measurements. CMV measurement Methods used in literature CMV specific IgG antibodies Measure virus activity Global measurement Virus detection with PCR Measures the virus presence locally Do not measure the activity Done in vitro Local CMV measurement No existing technique available Plan to develop measurement method: Find or develop a label that indicates CMV Attach radioactive or physical marker to the label Quantify and localise agent with nuclear (radioactive) or possibly magnetic resonance (physical) imaging technique A disadvantage of nuclear imaging is the exposure to radioactivity Plaque formation measurement (1) Not described or used in CMV-atherosclerosis studies It is assumed that there is no way to measure directly plaque growth Plaque growth can be determined by 2 separate plaque measurements Several possibilities are available to measure the amount of plaque locally (i.e. make an image of a subject that indicates plaque presence at each anatomical location) Plaque formation measurement (2) Plaque quantification techniques Epidemiological study Objective : Determine if a relation exists between CMV and plaque growth Two different possibilities Global level of CMV versus global amount of plaque growth Local level of CMV versus local amount of plaque formation CMV measurement Plaque measurement CMV level Global Local Total plaque progress Local Global Epidemiological study (2) Global CMV vs. global plaque Method: Three plaque measurements with MR imaging in carotid arteries combined with automatic plaque quantification Frequent CMV activity measurements in blood samples Population: a big random selection Determine number of necessary subjects from estimation of the probability to find CMV in plaque Determine: - Difference in plaque growth between the two periods (2plaque) - Amount of times the virus was activated (NCMV) Regression : 2plaque ~NCMV AbCMV Epidemiological study (3) Local CMV vs. local plaque Method Two plaque measurements with MR imaging in carotid arteries combined with automatic plaque quantification and localisation Image local CMV presence Computerised alignment of plaque and CMV images Population: a random selection Determine number of necessary subjects from estimation of the probability to find CMV in plaque and to find plaque Number of patients will be lower Determine: Plaque growth (plaque) at a large number of locations The presence of the virus at the same locations Regression : plaque ~ CMV CMV Local Plaque Local Summary and execution plan Molecular study design Epidemiological study design Molecular study Epidemiological study Hypothesis : CMV influences plaque formation More founded hypothesis Results ? Improved diagnosis New therapy

Slide 44 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computational modeling knowledge Research plan Define CMV measurement method Determine plaque formation quantification method Formalize epidemiological study Perform study in population Research plan Local vs. global measurements CMV measurement Plaque measurement Local measurements pinpoint the exact location of plaque and CMV (= histological) Global measurements (=systemic) indicate the total amount of plaque and CMV. It is less specific than local measurements. CMV measurement Methods used in literature CMV specific IgG antibodies Measure virus activity Global measurement Virus detection with PCR Measures the virus presence locally Do not measure the activity Done in vitro Local CMV measurement No existing technique available Plan to develop measurement method: Find or develop a label that indicates CMV Attach radioactive or physical marker to the label Quantify and localise agent with nuclear (radioactive) or possibly magnetic resonance (physical) imaging technique A disadvantage of nuclear imaging is the exposure to radioactivity Plaque formation measurement (1) Not described or used in CMV-atherosclerosis studies It is assumed that there is no way to measure directly plaque growth Plaque growth can be determined by 2 separate plaque measurements Several possibilities are available to measure the amount of plaque locally (i.e. make an image of a subject that indicates plaque presence at each anatomical location) Plaque formation measurement (2) Plaque quantification techniques Epidemiological study Objective : Determine if a relation exists between CMV and plaque growth Two different possibilities Global level of CMV versus global amount of plaque growth Local level of CMV versus local amount of plaque formation CMV measurement Plaque measurement CMV level Global Local Total plaque progress Local Global Epidemiological study (2) Global CMV vs. global plaque Method: Three plaque measurements with MR imaging in carotid arteries combined with automatic plaque quantification Frequent CMV activity measurements in blood samples Population: a big random selection Determine number of necessary subjects from estimation of the probability to find CMV in plaque Determine: - Difference in plaque growth between the two periods (2plaque) - Amount of times the virus was activated (NCMV) Regression : 2plaque ~NCMV AbCMV Epidemiological study (3) Local CMV vs. local plaque Method Two plaque measurements with MR imaging in carotid arteries combined with automatic plaque quantification and localisation Image local CMV presence Computerised alignment of plaque and CMV images Population: a random selection Determine number of necessary subjects from estimation of the probability to find CMV in plaque and to find plaque Number of patients will be lower Determine: Plaque growth (plaque) at a large number of locations The presence of the virus at the same locations Regression : plaque ~ CMV CMV Local Plaque Local Summary and execution plan Molecular study design Epidemiological study design Molecular study Epidemiological study Hypothesis : CMV influences plaque formation More founded hypothesis Results ? Improved diagnosis New therapy Imaging techniques Epidemiology Literature Biological understanding Task distribution Microarray data Medical importance Algorithms Team Pathway Interactions

Slide 45 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computational modeling knowledge Research plan Define CMV measurement method Determine plaque formation quantification method Formalize epidemiological study Perform study in population Research plan Local vs. global measurements CMV measurement Plaque measurement Local measurements pinpoint the exact location of plaque and CMV (= histological) Global measurements (=systemic) indicate the total amount of plaque and CMV. It is less specific than local measurements. CMV measurement Methods used in literature CMV specific IgG antibodies Measure virus activity Global measurement Virus detection with PCR Measures the virus presence locally Do not measure the activity Done in vitro Local CMV measurement No existing technique available Plan to develop measurement method: Find or develop a label that indicates CMV Attach radioactive or physical marker to the label Quantify and localise agent with nuclear (radioactive) or possibly magnetic resonance (physical) imaging technique A disadvantage of nuclear imaging is the exposure to radioactivity Plaque formation measurement (1) Not described or used in CMV-atherosclerosis studies It is assumed that there is no way to measure directly plaque growth Plaque growth can be determined by 2 separate plaque measurements Several possibilities are available to measure the amount of plaque locally (i.e. make an image of a subject that indicates plaque presence at each anatomical location) Plaque formation measurement (2) Plaque quantification techniques Epidemiological study Objective : Determine if a relation exists between CMV and plaque growth Two different possibilities Global level of CMV versus global amount of plaque growth Local level of CMV versus local amount of plaque formation CMV measurement Plaque measurement CMV level Global Local Total plaque progress Local Global Epidemiological study (2) Global CMV vs. global plaque Method: Three plaque measurements with MR imaging in carotid arteries combined with automatic plaque quantification Frequent CMV activity measurements in blood samples Population: a big random selection Determine number of necessary subjects from estimation of the probability to find CMV in plaque Determine: - Difference in plaque growth between the two periods (2plaque) - Amount of times the virus was activated (NCMV) Regression : 2plaque ~NCMV AbCMV Epidemiological study (3) Local CMV vs. local plaque Method Two plaque measurements with MR imaging in carotid arteries combined with automatic plaque quantification and localisation Image local CMV presence Computerised alignment of plaque and CMV images Population: a random selection Determine number of necessary subjects from estimation of the probability to find CMV in plaque and to find plaque Number of patients will be lower Determine: Plaque growth (plaque) at a large number of locations The presence of the virus at the same locations Regression : plaque ~ CMV CMV Local Plaque Local Summary and execution plan Molecular study design Epidemiological study design Molecular study Epidemiological study Hypothesis : CMV influences plaque formation More founded hypothesis Results ? Improved diagnosis New therapy Imaging techniques Epidemiology Literature Biological understanding Task distribution Microarray data Medical importance Algorithms Team Pathway Interactions Building the monastery of Les Avellanes, every family marked the stones because they were afraid that someone could steal them. We come also from different scientific families but respected the different point of views. We synergically built our cathedral project! What we learnt !

Slide 46 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computational modeling knowledge Research plan Define CMV measurement method Determine plaque formation quantification method Formalize epidemiological study Perform study in population Research plan Local vs. global measurements CMV measurement Plaque measurement Local measurements pinpoint the exact location of plaque and CMV (= histological) Global measurements (=systemic) indicate the total amount of plaque and CMV. It is less specific than local measurements. CMV measurement Methods used in literature CMV specific IgG antibodies Measure virus activity Global measurement Virus detection with PCR Measures the virus presence locally Do not measure the activity Done in vitro Local CMV measurement No existing technique available Plan to develop measurement method: Find or develop a label that indicates CMV Attach radioactive or physical marker to the label Quantify and localise agent with nuclear (radioactive) or possibly magnetic resonance (physical) imaging technique A disadvantage of nuclear imaging is the exposure to radioactivity Plaque formation measurement (1) Not described or used in CMV-atherosclerosis studies It is assumed that there is no way to measure directly plaque growth Plaque growth can be determined by 2 separate plaque measurements Several possibilities are available to measure the amount of plaque locally (i.e. make an image of a subject that indicates plaque presence at each anatomical location) Plaque formation measurement (2) Plaque quantification techniques Epidemiological study Objective : Determine if a relation exists between CMV and plaque growth Two different possibilities Global level of CMV versus global amount of plaque growth Local level of CMV versus local amount of plaque formation CMV measurement Plaque measurement CMV level Global Local Total plaque progress Local Global Epidemiological study (2) Global CMV vs. global plaque Method: Three plaque measurements with MR imaging in carotid arteries combined with automatic plaque quantification Frequent CMV activity measurements in blood samples Population: a big random selection Determine number of necessary subjects from estimation of the probability to find CMV in plaque Determine: - Difference in plaque growth between the two periods (2plaque) - Amount of times the virus was activated (NCMV) Regression : 2plaque ~NCMV AbCMV Epidemiological study (3) Local CMV vs. local plaque Method Two plaque measurements with MR imaging in carotid arteries combined with automatic plaque quantification and localisation Image local CMV presence Computerised alignment of plaque and CMV images Population: a random selection Determine number of necessary subjects from estimation of the probability to find CMV in plaque and to find plaque Number of patients will be lower Determine: Plaque growth (plaque) at a large number of locations The presence of the virus at the same locations Regression : plaque ~ CMV CMV Local Plaque Local Summary and execution plan Molecular study design Epidemiological study design Molecular study Epidemiological study Hypothesis : CMV influences plaque formation More founded hypothesis Results ? Improved diagnosis New therapy Imaging techniques Epidemiology Literature Biological understanding Task distribution Microarray data Medical importance Algorithms Team Pathway Interactions Building the monastery of Les Avellanes, every family marked the stones because they were afraid that someone could steal them. We come also from different scientific families but respected the different point of views. We synergically built our cathedral project! What we learnt ! Thank You Merci Gràcies Gracias Grazie Dank u Eskerrik asko

Slide 47 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computational modeling knowledge Research plan Define CMV measurement method Determine plaque formation quantification method Formalize epidemiological study Perform study in population Research plan Local vs. global measurements CMV measurement Plaque measurement Local measurements pinpoint the exact location of plaque and CMV (= histological) Global measurements (=systemic) indicate the total amount of plaque and CMV. It is less specific than local measurements. CMV measurement Methods used in literature CMV specific IgG antibodies Measure virus activity Global measurement Virus detection with PCR Measures the virus presence locally Do not measure the activity Done in vitro Local CMV measurement No existing technique available Plan to develop measurement method: Find or develop a label that indicates CMV Attach radioactive or physical marker to the label Quantify and localise agent with nuclear (radioactive) or possibly magnetic resonance (physical) imaging technique A disadvantage of nuclear imaging is the exposure to radioactivity Plaque formation measurement (1) Not described or used in CMV-atherosclerosis studies It is assumed that there is no way to measure directly plaque growth Plaque growth can be determined by 2 separate plaque measurements Several possibilities are available to measure the amount of plaque locally (i.e. make an image of a subject that indicates plaque presence at each anatomical location) Plaque formation measurement (2) Plaque quantification techniques Epidemiological study Objective : Determine if a relation exists between CMV and plaque growth Two different possibilities Global level of CMV versus global amount of plaque growth Local level of CMV versus local amount of plaque formation CMV measurement Plaque measurement CMV level Global Local Total plaque progress Local Global Epidemiological study (2) Global CMV vs. global plaque Method: Three plaque measurements with MR imaging in carotid arteries combined with automatic plaque quantification Frequent CMV activity measurements in blood samples Population: a big random selection Determine number of necessary subjects from estimation of the probability to find CMV in plaque Determine: - Difference in plaque growth between the two periods (2plaque) - Amount of times the virus was activated (NCMV) Regression : 2plaque ~NCMV AbCMV Epidemiological study (3) Local CMV vs. local plaque Method Two plaque measurements with MR imaging in carotid arteries combined with automatic plaque quantification and localisation Image local CMV presence Computerised alignment of plaque and CMV images Population: a random selection Determine number of necessary subjects from estimation of the probability to find CMV in plaque and to find plaque Number of patients will be lower Determine: Plaque growth (plaque) at a large number of locations The presence of the virus at the same locations Regression : plaque ~ CMV CMV Local Plaque Local Summary and execution plan Molecular study design Epidemiological study design Molecular study Epidemiological study Hypothesis : CMV influences plaque formation More founded hypothesis Results ? Improved diagnosis New therapy Imaging techniques Epidemiology Literature Biological understanding Task distribution Microarray data Medical importance Algorithms Team Pathway Interactions Building the monastery of Les Avellanes, every family marked the stones because they were afraid that someone could steal them. We come also from different scientific families but respected the different point of views. We synergically built our cathedral project! What we learnt ! Thank You Merci Gràcies Gracias Grazie Dank u Eskerrik asko CMV-Atherosclerosis from a medicinal chemistry point of view: rational multitarget polypharmacy STATINS Lipid alteration: Cholesterol-lowering properties Anti-inflammatoty effects: HMG-CoA Reductase inihibition No prenylated proteins to activate NF-kB. NF-kB is implicated in viral replication, inflammation, apoptosis and autoimmune disease.

Slide 48 -

Get insights into the relation between CMV and atherosclerosis Case study 1 Background Atherosclerosis The main cause of death in the Western world. Multi-factorial disease : environmental (diet, smoking, exercise, infection) and genetic risk factors. Inflammation is a main contributor. Is a disease in which a fatty-like substance (plaque) is deposited on the inside of the arteria walls. The exact cause of atherosclerosis remains unknown. Atherosclerosis and cytomegalovirus CMV is a double-stranded DNA beta herpes virus affecting 50 % of the population. Increasing evidence linking CMV and atherosclerosis (epidemiologic, antibodies, gene expression etc…) Nevertheless cellular mechanisms are not well understood and it is unknown whether the virus can causally contribute to atherosclerosis. Recently, lipid modulation experiments (statins, polyunsaturated fatty acid etc…), showed a potential mechanism in viral and inflammatory process. Aim The relation between CMV and atherosclerosis is controversial and not well understood. Therefore we aim to get insights into this relation Step 1 : To determine the proven relations between CMV and atherosclerosis Review of literature Results All studies focused on the relation between CMV and consequences of atherosclerosis (e.g. death and restenosis) Most studies conclude a positive relation between the two No studies focused on the process of plaque formation A negative relation has been demonstrated between CMV and plaque rapture Hypothesis: CMV influences plaque formation Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computarized modeling knowledge Cristina Medical Chemistry Michiel Computer Scientist Lula Biostatistician Adrián Computational Biologist Mathieu Biologist THE TEAM !!! Molecular Approach CMV Cell Network interaction Atherosclerosis Aim: Study the interactions between the CMV and the development of atherosclerosis Hypothesis: CMV trigger the formation of the plaques Molecular mechanisms are not well understood: implication of inflammatory/ immune response and lipids pathway Understand Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Simulation Parameter Estimation Sensitivity analysis Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Lipids Genes Lipid Metabolites Pro-Atherosclerotic genes Genes list of the key genes/metabolite involved in each process Literature Knowledge Define interactions between this emtities to build a network Text Mining Phase I : Building a Network of interactions genes 26 lipid genes related 11 pro-atherosclerotic genes 5 lipid Metabolites Text Mining tool : Pathway Studio (Use of options : add small molecules to integrate lipids). 99 proteins in total 98 small molecules (20 lipids) 771 interactions Text-Mining Pathway Assist network interaction of lipid metabolite / lipid genes / Atherosclerosis genes Limitations Text Mining Relation not always biologically relevant or true Not useful too many putative interactions Biological Interactions Metabolites / Genes are very limited Not useful too many putative interactions Computational Format compatibility Lack of quantitative relationship Difficult to work with this network Using this network of interactions we decided to build a more specialised and reduced network around lipid genes/metabolites and pro-atherosclerotic genes which could be use by biologists and computer scientists Creation of a curated and specialised interaction genes network Limitations to the direct (genes/proteins) interactions. Each interactions has been checked by one of us in the litterature and with our personal knowledge. Add manually Interactions with lipids (cholesterol, oxysterols, fatty acid). 19 proteins 3 lipids 65 interactions Pubmed GEO Array express 11 studies and dataset available statin GPX Time point experiments Different cell line (fibroblast, macrophages Species: murine, human 11 studies and dataset available CMV infection CMV/Lipids Database 31 Array experiments 9 studies and dataset on lipid modulation Genes Network + Metabolite Interactions model Micro-Array Database Validation and upgrading of the network Model Identifications of key genes and metabolites playing a role in atherosclerosis regulated by lipids regulation and or CMV infections. Pharmacology Computer science Molecular biology/ Bioinformatic Drugs targeting New Hypothesis on possible molecular interaction and bench validation Phase I Phase II Phase III Molecular Approach Strategy Tools Simulation Parameter Estimation Sensitivity analysis Tools for Computational biology Quantitative relationships between network nodes allow to work with different prediction and simulation software programs. Today, SBML is supported by over 90 software systems Today, SBML is supported by over 90 software systems No export support, we did it manually !!! Development of a computational tool !!! Logical translation of the pathway - positive interaction - negative interaction Power Laws: Applied in large datasets of metabolism. Not much biological details needed First insights into biochemical mechanism Mathematical method Rate Laws: Not enough knowledge Boolean Networks: Far from biochemical mechanism PDEs: Complicate method, few tools Simulation Results GDS476: 12 temporal points of 12626 genes PARAMETER DETERMINATION OF THE NETWORK Experimentally tunned model Parameter optimized model + Parameter optimized model + Sensitive node Pharmacological target Tool for biology Analysis of micro-Array data using our model to identify new target Clustering Pathway mapping Design of new micro-array experiments CMV Lipid modulation Host/CMV chip array Hypothesis key relations study them in a more “traditional” way Molecular biology Bench work Develop high throughput methods to record different lipids level into the cells Use of Mass spectrometry, development of lipids Array. Tool for pharmacology Identification of drugs targeting key genes and metabolites involved in the intersection between lipid and inflammatory pathways in a CMV / atherosclerosis context. CMV infection is involved in two of the major mechanism that lead development of atherosclerosis: CMV + + Immune injury Lipid Alteration CMV - - CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy STATINS LXRα and PPARα agonist Use of these drugs to limit CMV effects in atherosclerosis disease : lipid alteration and inflammatory effect CMV-Atherosclerosis from a medicinal chemistry point of view: rational multi-target poly-pharmacy Antagonist of Nf-kB to reduce immunity response through: inhibition of iNOs genes. It produces nitric oxide (NO) that increase oxLDL Inhibition of leukocyte adhesion cascade NFKB expression is increase by CMV. PPAR-gamma agonists induced expression of ABCG1. ABCG1 has been shown to transfer cholesterol from cells to form HDL, the carrier of “good cholesterol” in the blood and is regulated by CMV. PPAR-alpha agonist induce expression of LXRα. CMV-Atherosclerosis from a medicinal chemistry point of view: dual PPARα and PPARγ agonist Strategy Aim : to get insights into the relation between CMV and atherosclerosis Review of literature Hypothesis: CMV influences plaque formation Design of medical strategy New epidemiological study design Imaging Pharmacology Design of molecular study Model MicroArray Computational modeling knowledge Research plan Define CMV measurement method Determine plaque formation quantification method Formalize epidemiological study Perform study in population Research plan Local vs. global measurements CMV measurement Plaque measurement Local measurements pinpoint the exact location of plaque and CMV (= histological) Global measurements (=systemic) indicate the total amount of plaque and CMV. It is less specific than local measurements. CMV measurement Methods used in literature CMV specific IgG antibodies Measure virus activity Global measurement Virus detection with PCR Measures the virus presence locally Do not measure the activity Done in vitro Local CMV measurement No existing technique available Plan to develop measurement method: Find or develop a label that indicates CMV Attach radioactive or physical marker to the label Quantify and localise agent with nuclear (radioactive) or possibly magnetic resonance (physical) imaging technique A disadvantage of nuclear imaging is the exposure to radioactivity Plaque formation measurement (1) Not described or used in CMV-atherosclerosis studies It is assumed that there is no way to measure directly plaque growth Plaque growth can be determined by 2 separate plaque measurements Several possibilities are available to measure the amount of plaque locally (i.e. make an image of a subject that indicates plaque presence at each anatomical location) Plaque formation measurement (2) Plaque quantification techniques Epidemiological study Objective : Determine if a relation exists between CMV and plaque growth Two different possibilities Global level of CMV versus global amount of plaque growth Local level of CMV versus local amount of plaque formation CMV measurement Plaque measurement CMV level Global Local Total plaque progress Local Global Epidemiological study (2) Global CMV vs. global plaque Method: Three plaque measurements with MR imaging in carotid arteries combined with automatic plaque quantification Frequent CMV activity measurements in blood samples Population: a big random selection Determine number of necessary subjects from estimation of the probability to find CMV in plaque Determine: - Difference in plaque growth between the two periods (2plaque) - Amount of times the virus was activated (NCMV) Regression : 2plaque ~NCMV AbCMV Epidemiological study (3) Local CMV vs. local plaque Method Two plaque measurements with MR imaging in carotid arteries combined with automatic plaque quantification and localisation Image local CMV presence Computerised alignment of plaque and CMV images Population: a random selection Determine number of necessary subjects from estimation of the probability to find CMV in plaque and to find plaque Number of patients will be lower Determine: Plaque growth (plaque) at a large number of locations The presence of the virus at the same locations Regression : plaque ~ CMV CMV Local Plaque Local Summary and execution plan Molecular study design Epidemiological study design Molecular study Epidemiological study Hypothesis : CMV influences plaque formation More founded hypothesis Results ? Improved diagnosis New therapy Imaging techniques Epidemiology Literature Biological understanding Task distribution Microarray data Medical importance Algorithms Team Pathway Interactions Building the monastery of Les Avellanes, every family marked the stones because they were afraid that someone could steal them. We come also from different scientific families but respected the different point of views. We synergically built our cathedral project! What we learnt ! Thank You Merci Gràcies Gracias Grazie Dank u Eskerrik asko CMV-Atherosclerosis from a medicinal chemistry point of view: rational multitarget polypharmacy STATINS Lipid alteration: Cholesterol-lowering properties Anti-inflammatoty effects: HMG-CoA Reductase inihibition No prenylated proteins to activate NF-kB. NF-kB is implicated in viral replication, inflammation, apoptosis and autoimmune disease. CMV-Atherosclerosis from a medicinal chemistry point of view: rational multitarget polypharmacy LXRα and PPARα agonist LXRα agonists: decrease circulating LDL issue cholesterol increase HDL but hypertriglyceridemia PPARα agonists: activates β-oxidation in the liver increase HDL cholesterol PPARα agonists inhibit iNOS expression. NO is an inflammatory target. Lipid alteration: Anti-inflammatory effect: